Nuclear translocation of PKM2 modulates astrocyte proliferation via p27 and β-catenin pathway after spinal cord injury

Abstract

Aberrant functionality of the cell cycle has been implicated in the pathology of traumatic SCI. Although it has been reported that the expressions of various cell cycle related proteins were altered significantly following SCI, detailed information on the subject remains largely unclear. The embryonic pyruvate kinase M2 (PKM2) is an important metabolic kinase in aerobic glycolysis or the warburg effect, however, its functions in central nervous system (CNS) injury remains elusive. Here we demonstrate that PKM2 was not only significantly upregulated by western blot and immunohistochemistry but certain traumatic stimuli also induced translocation of PKM2 into the nucleus in astrocytes following spinal cord injury (SCI). Furthermore, the expression levels and localization of p-β-catenin, p27, cyclin D1 and PCNA were correlated with PKM2 after SCI. In vitro, we also found that PKM2 co-immunoprecipitation with p-β-catenin and p27 respectively. Knockdown of PKM2 apparently decreased the level of PCNA, cyclinD1, p27 in primary astrocyte cells. Taken together, our findings indicate that nuclear translocation of PKM2 promotes astrocytes proliferation after SCI through modulating cell cycle signaling. These discoveries firstly uncovered the role of PKM2 in spinal cord injury and provided a potential therapeutic target for CNS injury and repair.

Abbreviations

CNS

Central nervous system

CDKs

cyclin-dependent kinases

CAK

CDK activating kinase

E2F

E2 promoter-binding protein dimerization partners

GFAP

Glial fibrillary acidic protein

GAPDH

Glyceraldehyde-3-phosphate dehydrogenase

NeuN

Neuronal nuclei

PCNA

Proliferating cell nuclear antigen

PEP

phosphoenolpyruvate

PK

Pyruvate kinase

PKM2

pyruvate kinase M2

pRb

protein retinoblastoma

SCI

Spinal cord injury

EGFR

Epidermal growth factor receptor.

Introduction

Traumatic spinal cord injury (SCI) is one of the major causes of irreversible nerve injury, leading to both tissue loss and associated neurological dysfunction.¹ Biochemical processes are initiated in the hours to weeks after SCI that further damage the tissue within and surrounding the initial injury site: a process termed secondary injury, which is one of the critical contributing factors of SCI pathology.² The injury at the initial lesion site of impact is exacerbated by this process, which also results in spreading of the damage to adjacent tissue throughout the spinal cord.

The secondary injury is caused by ischemia/reperfusion, electrolyte imbalance, free radical damage, ischemia, excitotoxicity, apoptosis, and inflammation,^3-6 which all contribute to neuronal cell death, astrocyte activation and proliferation.^7,8 Rapid proliferation of astrocytes is benefited from abnormal modulation of cell cycle. Overactivation of astrocytes can result in the formation of a dense astrocytic scar.⁹ This glial scar acts as a physical and biochemical barrier to regeneration and plasticity due to the expression of inhibitory components, and affects functional recovery from SCI.^10,11 Thus, it seems to be crucial for the treatment of SCI to investigate the molecular mechanisms of glial cell activation and regulate abnormal cell cycle by which they can be regulated suitably during the repair process after SCI.

Cell cycle inhibition has been shown to reduce glial proliferation and scar formation after traumatic injury,¹² while cell cycle activation contribute to post-mitotic cell death, glial cell activation and proliferation after SCI.^13,14 Among G1 to S phase and the commitment of a cell to progress, a group of proteins called cyclins regulate tightly the activity of cyclin-dependent kinases (CDKs).¹⁵ The activity of G1-phase cyclin CDKs is also negatively regulated through the association of specific CDK inhibitors (CKIs), including the INK4 family proteins and the Cip/Kip family proteins p27^kip116. p27^Kip1 (p27) is a key cell-cycle regulator, which negatively regulates cell cycle progression via inhibiting the cyclinE/CDK2 complexes directly.^17,18 The activity of p27 is controlled by its concentration, distribution among different cellular complexes, and its cellular location.^19,20

The final step in glycolysis is catalyzed by Pyruvate kinase (PK) via transferring the phosphate from phosphoenolpyruvate (PEP) to ADP, thereby generating pyruvate and ATP.²¹ PK has 4 isoforms in mammals: PKL, PKR, PKM1, and PKM2, were expressed in various cells respectively. The PKM1 and PKM2 isoforms result from mutually exclusive alternative splicing of the PKM pre-mRNA that lead to inclusion of either exon 9 (PKM1) or exon 10 (PKM2).²² PKM2 is expressed in all proliferating cells such as embryonic cells, adult stem cells, and especially tumor cells. These cells present the common feature: high rates of nucleic acid synthesis.²³ In contrast to the established role of PKM2 in aerobic glycolysis or the warburg effect,^24-26 Epidermal growth factor receptor (EGFR) activation induces translocation of PKM2 into the nucleus to bind phosphorylated β-catenin. PKM2-dependent β-catenin transactivation is instrumental in EGFR-promoted tumor cell proliferation and brain tumor development. In this state, both proteins should be recruited to the cyclin D1(CCND1) promoter for this interaction, making HDAC3 removal from the promoter, histone H3 acetylation, and cyclin D1 expression.²⁷ Cyclin D1, a positive regulator of the cell cycle, can bind and modulate the actions of transcription factors and also function as a transcriptionl co-regulator in different cell types and particular in astroglial cells.^28-31 In this study, the expression of PKM2, p27, p-β-catenin, and cyclin D1 in the adult spinal cord of contusion injury rat was analyzed. To find new targets for the repair of SCI, our study examined the role of PKM2 in the cell proliferation in the spinal cord of SCI damage, which may have important significance in clinical therapy.

Materials and Methods

Primary astrocyte cell culture, siRNA, and transfection

Primary astrocytes were prepared from spinal cords of newborn Ninety Sprague Dawley (SD) rats (postnatal day 3) using a previously described method with some modifications.^32,33 Briefly, spinal cords of neonatal rats were ejected from the vertebral column, treated with 0.125% trypsin for 10 min, followed by mechanical trituration in complete Dulbecco's modified essential medium (DMEM + 10% heat-inactivated fetal bovine serum (FBS); Sigma). After centrifugation at 800 rpm for 5 min, the cells were suspended in DMEM with nutrient mixture F12 (1:1 v/v; Sigma), containing FBS and 1% penicillin/streptomycin, and plated in 6-well dishes. The cultures were maintained in an atmosphere of 95% O₂/5% CO₂ at 37°C. Approximately on day 10 and 11, oligodendrocytes and microglial cells growing on top of the confluent astrocyte layer were removed by shaking at 200 rpm for 2h at 37°C and replacing the culture medium. The next day, the cells were trypsinized and replated in 6-well plates (40,000 cells/well). The cultures were used for subsequent experiments when confluent (typically within 4 – 6 d).

For specific interference protein expression, the control and PKM2 siRNA oligos (Biomics Technology) were used. All of the transfection assays were performed with Lipofectamine 2000 transfection reagent (Invitrogen) according to the manufacturer's protocol.

Astrocyte cell immunoprecipitation

For specificity of interactions about PKM2, normal and Escherichia coli lipopolysaccharide (LPS) serotype 0111:B4 (Sigma) induced primary astrocytes were lysed with 1× cell lysis buffer (Cell Signaling) and the lysate was centrifuged. Astrocytes were stimulated by LPS (100 ng/mL) for 3, 6, 9, 12 and 24 hours. The supernatant was precleared with protein A/G beads (Sigma, USA), and incubated with indicated antibody overnight.. Thereafter protein A/G beads were applied, all at 4°C. After 2 h of incubation, pellets were washed 5 times with lysis buffer and resuspended in sample buffer and analyzed by SDS–PAGE.

Establishment of rat SCI model

All surgical interventions and postoperative animal care were carried out in accordance with the Guide for the Care and Use of Laboratory Animals [National Research Council, 1996, USA] and were approved by the Chinese National Committee to the Use of Experimental Animals for Medical Purposes, Jiangsu Branch. Male SD rats (n = 66) with an average body weight of 250 g (220–275 g) were used in this study. Rats were deeply anesthetized with chloral hydrate (10% solution) and dorsal laminectomies at the level of the ninth thoracic vertebra (T9) were carried out. Contusion injuries (n = 54) were performed using the NYU impactor,³⁴ the exposed spinal cord was contused by dropping a 10 g weight and 2.0-mm-diameter rod from a height of 75 mm. After SCI, the overlying muscles and skin were closed in layers with 4–0 silk sutures and staples, respectively, and the animals were allowed to recover on a 37°C heating pad. Postoperative treatments included saline (2.0 mL, s.c.) for rehydration and Baytril (0.3 mL, 22.7 mg/ml, s.c., twice daily) to prevent urinary tract infection. Bladders were manually expressed twice daily until emptying returned. Animals were sacrificed at 6 h, 12 h, 1 day, 3 days, 5 days, 7 days, and 14 d after injury, and at least 3 animals/experimental group were analyzed. Twelve sham animals were served as non-injuryed controls.

Microarray analysis

The total RNA was immediately extracted from naive and 5-day-injured spinal cord tissues using Trizol (Invitrogen) and then transcribed into cDNAs with random primers using the SuperScript II First-Strand Synthesis System (Invitrogen). The microarray assay was performed with the assistance of GeneTech Limit Co. (Shanghai, China). Briefly, the single-stranded cDNA was purified using a Qiagen PCR purification column (Qiagen, Valencia, CA) and fragmented with DNAse I (Sigma). The purified cDNAs were biotin-labeled using GeneChip DNA Labeling Reagent (Affymetrix, Santa Clara, CA) and directly hybridized to the GeneChip Rat Genome 230 2.0 array (Affymetrix). Then, the samples were incubated at 48°C in a GeneChip hybridization oven 640 (Affymetrix) for 16 hr. The microarrays were washed and stained with the GeneChip Fluidics Station 450 (Affymetrix). Finally, scanning was performed with a GeneChip Scanner 3000 using manufacturer-recommended default settings. The raw data were analyzed in Partek GS 6.5 software.

Western blot analysis

The procedures were carried out similarly to previously described methods²⁰ To prepare lysates, the 5 mm rostral and 5 mm caudal to the injury epicenter was removed, and frozen spinal cord samples were minced with eye scissors in ice. Then, the samples were then homogenized in lysis buffer (1% NP-40, 50 mmol/l Tris, pH 7.5, 5 mmol/l EDTA, 1% SDS, 1% sodium deoxycholate, 1% Triton X-100, 1 mmol/l PMSF, 10mg/ml aprotinin, and 1mg/ml leupeptin) and clarified by centrifuging for 20 min in a microcentrifuge at 4°C.

To obtain lysates from cultured primary astrocyte cells, the RIPA buffer (50 mM Tris, pH 7.4, 150 mM NaCl, 1% NP-40, 0.5% sodium deoxycholate, 0.1% SDS, and 1×complete protease inhibitor cocktail) was used to lyse transfected astrocyte cells. After determination of its protein concentration with the Bradford assay (Bio-Rad), the resulting supernatant was subjected to SDS-polyacrylamide gel electrophoresis (SDS-PAGE). The separated proteins were transferred to a polyvinylidene fluoride (PVDF; Millipore) membrane by a mini transblot (Bio-Rad). The membrane was then blocked with 5% nonfat milk and incubated with primary antibody against PKM2 (anti-rabbit, 1:500; Santa Cruz), p27 (anti-mouse, 1:500; Santa Cruz), p-β-catenin (anti-mouse, 1:1000; Santa Cruz), Cyclin D1 (anti-mouse, 1:1000; Santa Cruz), proliferating cell nuclear antigen (PCNA) (anti-mouse, 1:1000; Santa Cruz), GFAP, Tubulin, LaminB, CDK2, Cyclin E, CDK4, p-Rb, GAPDH (anti-rabbit and mouse, 1:1000; Santa Cruz) . After incubating with an anti-rabbit or anti-mouse horseradish peroxidase-conjugated secondary antibody, protein was visualized using an enhanced chemiluminescence system (ECL, Pierce Company, USA).

Preparation of transverse cryosections and immunohistochemistry

After defined survival times, sham and injured rats were terminally anesthetized and perfused through the ascending aorta with saline, followed by 4% paraformaldehyde (each time point, n = 3). After perfusion, the sham and injured spinal cords were removed and post-fixed in the same fixative solution for overnight at 4°C and immersed with 20% sucrose for 2–3 days, followed by 30% sucrose for 2–3 d After treatment with sucrose solutions, the tissues were embedded in O.C.T compound. Then 5-mm frozen cross-sections at 2 spinal cord levels (2 mm rostral and caudal to the epicenter of injury) were prepared and then examined.

All of the sections were blocked with 10% donkey serum with 0.3% Triton X-100 and 1% (w/v) bovine serum albumin (BSA) for 2 h at room temperature (RT) and incubated overnight at 4°C with an anti-PKM2 antibody (anti-rabbit, 1:500; Cell Signaling) and Phosphate Buffered Saline (PBS) instead of primary antibody for negative control, followed by incubation in biotinylated secondary antibody (Vector Laboratories, Burlin-game, CA). Staining was visualized with DAB (Vector Laboratories). Cells with strong or moderate brown staining were counted as positive, cells with no staining were counted as negative, and cells with weak staining were scored separately.

Double-immunofluorescence assay

For double-immunofluorescence assay, cryosections were blocked with a blocking buffer containing 10% donkey normal serum, 3% BSA, 0.1% Triton X-100, and 0.05% Tween-20 at 25°C for 2 h. Then, the sections were incubated with primary antibodies PKM2 (1:200; Santa Cruz Biotechnology), mouse monoclonal anti-NeuN (1:600; Millipore, Bedford, MA), mouse monoclonal anti-GFAP (1:200; Sigma), mouse or goat monoclonal anti-PCNA, anti-p27, anti-Cyclin D1, anti-p-β-catenin and Phosphate Buffered Saline (PBS) instead of primary antibody for negative control overnight at 4°C. Briefly, sections were incubated with both primary antibodies overnight at 4°C. Next, the sections were incubated with a mixture of FITC- and TRITC-conjugated secondary antibodies for 2 h at 25°C. The stained sections were visualized with a Leica confocal microscope.

Quantitative and statistical analysis

The numbers of PKM2-positive cells in the spinal cord 2 mm rostral to the epicenter were counted in a 500 μm × 500 μm measuring frame. For each animal, a measure was taken in a section through the dorsal horn, the lateral funiculus and the ventral horn. To avoid counting the same cell in more than one section, we counted every fifth section (50 μm apart). The cell counts were then used to determine the total number of PKM2-positive cells per square millimeter. Cells double labeled for PKM2 and the other phenotypic-specific markers used in the experiment were also quantified in each section. All data were analyzed with SPSS17.0 statistical software. All values were expressed as the mean ±SEM. One-way ANOVA followed by the Tukey's post-hoc multiple comparison tests and un-paired t test for double comparison were used for statistical analysis. P–values less than 0.05 were considered statistically significant.

Results

Temporal expression of PKM2 in rat spinal cord after SCI

To identify the differentially expressed genes in rat spinal cord after contusion injury, we established a penetrating SCI model with adult rats. The total mRNA from noninjured spinal cord and 5-day-injured spinal cord was extracted and subjected to microarray analysis. In this way, a gene encoding embryonic pyruvate kinase M2 was identified to be remarkably upregulated in injured spinal cord (data not shown). To examine whether PKM2 was altered in the process of SCI, Western blot analysis was performed using anti-PKM2 antibody. As shown in Figure 1A, B, the protein level of PKM2 was significantly increased at 1 day and reached a maximum 5 d following SCI.

Figure 1.

Western Blot analysis the expression profiles of PKM2 following SCI. Spinal cord tissues from rats at various survival times after SCI were homogenized and subjected to immunoblot analysis (A). Samples immunoblots probed for PKM2 and GAPDH are shown. The bar chart below demonstrates the ratio of PKM2 to GAPDH for each time point (B). The data are means ±SEM (n = 3, * P < 0.05, significantly different from the sham group).

The distribution of PKM2 in the spinal cord

To identify the distribution of PKM2 after SCI, we performed immunohistochemistry with anti-PKM2 monoclonal antibodies. As protein gel blot analysis shown, PKM2 have the highest protein expression at day 5 after SCI, so we chose day 5 as the time point of immunohistochemistry. In the rostral spinal cord 2 mm from the epicenter, PKM2 were extensively expressed in both the gray and white matter of spinal cord whether the animals were sham or injured (Fig. 2A, B). Notably, PKM2 was primarily expressed in the cytoplasm of the white matter in sham group (Fig. 2E), while PKM2 was transported into the nucleus and high expression in the nucleus at 5 day after injury (Fig. 2F). However, this biological phenomenon is not obvious in the gray matter (Fig. 2C, D). No staining was observed in the negative control sections (Fig. 2G).The profiles of spatial distribution of PKM2 were measured between sham group and day 5 after injury group. The data were consistent with western blot results (Fig. 2H). Both protein gel blot and immunohistochemical analysis indicate that contusion injuries of spinal cord induced up-regulation of PKM2 in vivo.

Figure 2.

Immunohistochemical analysis the distribution of PKM2 in the adult rat spinal cord. Low-power views of cross-sections immunostained with antibody specific for PKM2 in naive spinal cord (A) and day 5 after injury (B). Higher-power views in the gray (C, D) and white matter (E, F) of sham and injured spinal cord. PKM2 staining was mostly localized to the cytosol in white matter, while PKM2 was transported into nucleus and sequestered in the cytosol, nucleus staining was increased at day 5 after injury (F). This change was not obvious the gray matter (C, D). Phosphate Buffered Saline (PBS) instead of primary antibody for negative control (G). Scale bars: 200 μm (A, B) and 20 μm (C–G). The quantitative analysis of PKM2 positive cells/mm² between naive and day 5 after SCI. * Significant difference at P < 0.05 compared with naive. Error bars represent SEM.

The colocalization of PKM2 with different phenotype-specific markers in spinal cord

To further address the expressions of PKM2 in spinal cord, immunofluorescent staining was performed with the following cell-specific markers: GFAP (astrocyte marker) and NeuN (neuron marker). As shown in Figure 3, PKM2 were expressed in astrocytes and neurons at noninjured spinal cord and 5-day-injured spinal cord (Fig. 3A–L). Interestingly, PKM2 expression was increased more significantly in astrocytes at day 5 after SCI compared with sham spinal cord (Fig. 3A–F), and less significantly in neurons (Fig. 3G–L). In addition, PKM2 was high expression in the nucleus at day 5 after SCI. No staining was observed in the negative control sections (Fig. 3M, N). These findings suggested that the altered expression of PKM2 occurred mainly in astrocytes in the process of SCI, implying that PKM2 might contribute to the physiological changes in spinal cord astrocytes following injury.

Figure 3.

Double immunofluorescence staining for PKM2 with GFAP, NeuN in spinal cord. In the adult rat spinal cord sham and 2 mm to epicenter at day 5 after SCI, horizontal sections labeled with PKM2 (red) with different phenotype-specific markers (green), including GFAP(astrocytic marker, A–F), NeuN (neuronal marker, G–L) and the co-localization (yellow, arrows). Phosphate Buffered Saline (PBS) instead of primary antibody for negative control (M, N).The quantitative analysis of different phenotype-specific markers positive cells expressing PKM2 (%) in the sham spinal cord and day 5 after SCI. The changes of PKM2 positive expression cells after SCI were prominent in the astrocytes. Significant difference at P < 0.05 compared with naive animals. The error bars represent SEM. Scale bars: 20 μm.

Correlation between altered expressions of PKM2 with pathological alterations of astrocytes following SCI

To examine whether the increased expression of PKM2 was associated with astrocyte proliferation during SCI, we examined cell cycle pathway-related proteins changes after contusion injury. As shown in Fig. 4A, B, the expression of PCNA, Cyclin D1 and p-β-catenin was correlated with the protein level of PKM2 after SCI. In contrast, the endogenous CDK inhibitor p27 was inversely correlated with the expression of PKM2. Furthermore immunofluorescent analysis was performed on transverse cryosections with anti-PKM2, GFAP, PCNA, cyclin D1, p27, and p-β-catenin antibodies (Fig. 4C–R). PKM2 with PCNA, cyclin D1, p27, and p-β-catenin was coexpressed in GFAP-positive astrocytes at day 5 after spinal cord injury. The results indicated that the changes in PKM2 expression might be associated with cell cycle progression of astrocytes stimulated by spinal cord injury.

Figure 4.

Association of PKM2 with proliferation after SCI. Sample immunoblots probed for PCNA, p27, p-β-catenin and cyclin D1 in the spinal cord after injury and GAPDH were shown above. PCNA, p-β-catenin and cyclin D1 expression was obviously increased at days 3–7 after SCI and maintained for 2 weeks, while p27 expression was inversely (A). Semiquantitative analysis (relative density) of the intensity of the proteins to GAPDH at the indicated time points (B). The data are mean ±SEM (n = 3;* P < 0.05, significantly different from the sham groups). C-R: Co-localization immunofluorescence staining for PKM2 (red), GFAP (green), PCNA, cyclin D1, p27 and p-β-catenin (purple) after SCI. There were co-localizations between PKM2 and PCNA (O), cyclin D1 (P), p27 (Q) and p-β-catenin (R) in reactive astrocytes (GFAP-positive). Scale bars: 20 μm.

Expression of PKM2 in LPS-induced proliferative primary astroglia cells

To further investigate PKM2 function was related to astrocyte proliferation, we used the model of proliferation of astrocyte cultured in vitro. Astrocytes were stimulated by LPS (100ng/mL) to mimic spinal cord injury stimulation to astrocytes. The expression of PKM2 was immediately increased after LPS treatment, and the level of GFAP and PCNA was significantly increased, suggesting that PKM2 was related to proliferation following LPS exposure (Fig. 5A, B). Previous studies showed that PKM2 was high expression in the nucleus following SCI (Figs. 2 and 3). We further examined the location of PKM2 in the primary astroglia cells after LPS stimulation in vitro. As predicted, the expression PKM2 in nucleus were upregulated at 12 hr following LPS exposure, and the cytoplasm expression was decreased (Fig. 5C). To further elucidate the translocation of PKM2, a PKM2 specific siRNA was used (Fig. 5D, E). Furthermore, immunofluorescent analysis was performed on astrocyte cells after LPS stimulation. Similarly, PKM2 seemed to be mainly in the nucleus after administration of LPS for 12 hr. However, the groups of control and PKM2 siRNA were unchanged (Fig. 5F-Q). These data suggested that LPS stimuli of astrocyte cells in vitro also induced upregulation of PKM2 in a time-dependent manner.

Figure 5.

Analysis of PKM2 expression during the proliferation of primary astrocyte cells. Primary astrocyte cells were stimulated by LPS for 24 h. Western blots showed PKM2 was significantly increased after LPS stimulation as well as PCNA and GFAP (A) The bar chart below demonstrates the ratio of PKM2, GFAP and PCNA to GAPDH for each time point (B). The data are means ± SEM (n = 3,* P < 0.05, significantly different from the sham group). Nuclear and cytosolic proteins were immunoblotted for PKM2 (C). The levels of lamin B and tubulin in the nuclear and cytosolic fractions were also immunoblotted to confirm the purity of the subcellular fractions, respectively. Primary astrocyte cells were mock transfected or transfected with nonspecific siRNA, or PKM2 siRNA oligos were subjected to protein gel blot analysis with PKM2 and GAPDH antibodies at 48 hr after transfection (D). Quantitative analysis of the intensity of PKM2 to GAPDH (E). Astrocyte cells were mock transfected or transfected with nonspecific siRNA, or PKM2 siRNA oligos in the presence of LPS for 24 h. After treatment, PKM2 was largely redistributed to the nucleus and the amount in the cytoplasm declined in mock and nonspecific siRNA groups (F-Q). The data are mean ±SEM (n = 3;* P < 0.05, significantly different from the control groups).

Function of PKM2 on astrocyte cells proliferation

PKM2 can translocate into the nucleus to bind phosphorylated β-catenin in U87/EGFR human glioblastoma (GBM) cells. Thus, we probed whether the specificity of interactions between PKM2 and p-β-catenin also happened in astrocytes of spinal cord injury. Immunoprecipitation studies were performed to demonstrate complex formation between PKM2 and p-β-catenin in cultured primary astrocytes in vitro (Fig. 6A). Indeed, PKM2 could be precipitated with p-β-catenin, indicating that these 2 proteins were in the same complex in astrocytes after spinal cord injury. Recent data show the cyclin D1 expression was regulated by the interaction of PKM2 and p-β-catenin.²⁷ To further elucidate the change of cyclin D1 induced by PKM2 and p-β-catenin in the regulation of astrocyte cells proliferation, we assessed the expression of cyclin D1, CDK4 and p-Rb in astrocyte cells after LPS. We observed that up-regulated cyclin D1, CDK4 and p-Rb were positively correlated with PKM2 and p-β-catenin by western blot analysis (Fig. 6B, C). Furthermore, p-β-catenin was down-regulated after PKM2 specific siRNA (Fig. 6G, H). These data indicate that PKM2 can interact with p-β-catenin and further promote astrocyte cells proliferation via regulating cell cycle related proteins.

Figure 6.

Analysis of PKM2 functions during the proliferation of primary astrocyte cells. Co-immunoprecipitation analysis of interactions of p-β-catenin and p27 with PKM2 in spinal cord (A, D). Samples were subjected to immunoprecipitation using antibody against PKM2, and PKM2 were co-immunoprecipitated with p-β-catenin and p27. Moreover, the associations of p-β-catenin and p27 with PKM2 were evidently increased at day 5 after SCI. IP indicates samples subjected to IP procedure; input indicates positive control; IgG is used to exclude specific binding. Astrocyte cells were collected at the times indicated after LPS stimulation. P-β-catenin, cyclin D1, CDK4 and p-Rb, p27, CDK2 and cyclin E expression were evaluated by western blotting respectively after LPS stimulation (B, E). Semiquantitative analysis (relative density) of the intensity of the proteins to GAPDH at the indicated time points (C, F). The data are mean ±SEM (n = 3; *P < 0.05, significantly different from the naive groups). G: Astrocyte cells were transfected with nonspecific siRNA, or PKM2 siRNA were subjected to Western blot analysis with p27, cyclin D1 and PCNA antibodies at 48 h after transfection. Semiquantitative analysis (relative density) of the intensity of the proteins to GAPDH at the indicated time points (H).The data are mean ±SEM (n = 3; *P < 0.05, significantly different from the nonspecific siRNA groups).

Abnormal positive modulation of cell cycle accelerates rapid proliferation of astrocytes. p27 is a key cell-cycle regulator, so we also observed that whether PKM2 interact with p27 in astrocytes. Interestingly, we found the positive results (Fig. 6D), indicating PKM2 may also participate in modulation of cell cycle by interacting with p27. Furthermore, p27 was upregulated following PKM2 specific siRNA (Fig. 6G, H). Meanwhile, we detected the expression of cyclin E and CDK2 in astrocyte cells after LPS. Western blot analysis show cyclin E and CDK2 were up-regulated (Fig. 6E, F), which correlated with cyclin D1, CDK4 and p-Rb. All of these indicate PKM2 is involved in the astrocyte proliferation after spinal cord injury, and the mechanism of PKM2 functions might rely on the interact with p-β-catenin and p27 in modulating cell cycle signaling.

Discussion

Traumatic SCI is a crushing disorder and a great financial burden on society and family. It is a favorable manner to investigate the cellular and molecular mechanisms involved in the secondary insult by using the model of rat spinal cord contusion injury to recapitulate the clinical scenario.³⁵ In vivo study, we demonstrated PKM2 expression was increase and positively correlated with β-catenin and cyclin D1 expression, but inversely associated with p27 expression. Additionally, double immunofluorescence staining showed coexpression of PKM2 with β-catenin, p27, cyclin D1, and PCNA were predominant in astrocyte cells, which were largely proliferating. More interestingly, the change location of PKM2 prompted certain traumatic stimuli can induce translocation of PKM2 into the nucleus following spinal cord injury. The data provided a preliminary evidence of PKM2 functions underlying CNS injury and repair.

The secondary SCI includes complex physiological and biochemical mechanisms.³⁶ An important mechanism in the pathophysiology of SCI is the activation of astrocyte proliferation, characterized by increased expression of the astrocyte-specific marker GFAP.^35,37 Although astrocytes secrete critical growth factors for neurons, the glial scar presents a physical barrier to regeneration and plasticity.^38,39 It is a source of multiple inhibitory factors that may limit neuroplasticity. Thus, a better control of astrocytes proliferation after SCI and identify of proteins involved in this process will be helpful in aiding the regeneration of the spinal cord following injury.

It has been demonstrated EGFR activation induces translocation of PKM2 into the nucleus, where K433 of PKM2 binds to c-Srcphosphorylated Y333 of β-catenin.⁴⁰ This interaction is required for both proteins to be recruited to the CCND1 promoter, leading to HDAC3 removal from the promoter, histone H3 acetylation, and cyclin D1 expression.²⁷ During the traumatic SCI, PKM2 was highly expressed in reactive astrocytes of spinal cord and translocation into the nucleus. Meanwhile, PKM2 co-labeled with β-catenin in GFAP positive cells and interacted with β-catenin in primary astroglia cells. During LPS treatment, PKM2 level was increasing accompanied with increasing β-catenin and cyclin D1, which is consistent with the data in vitro. We stimulated the PKM2 specific siRNA-transfected astrocyte cells and found the decreased level of p-β-catenin at protein level. Based on these, we conjectured that PKM2 might be associated with astrocyte proliferation via regulating β-catenin and cyclin D1 expression.

During the spinal cord injury and repair process, cell cycle regulation may be indispensable. As a member of Cip/Kip family of cyclin dependent kinase inhibitors, p27 exerts its function on cell cycle regulation and neurogenesis in the developing CNS.²⁰ The loss of p27 leads to an increase in cell proliferation, while overexpression of p27 arrests cells in G1 phase.⁴¹ Recently reports have showed p27 was downregulated in the injured thoracic spinal cord⁴² and the lumbar spinal cord after sciatic nerve injury,⁴³ and the downregulation of p27 was very likely to be involved in the nervous system injury and repair process after SCI. Thus, the amount of p27 expression is very important for the injury and repair of spinal cord. p27 negatively regulates cell cycle progression by directly inhibiting the cyclin E/CDK2 complexes in mammalian cells.¹⁷ In the current study, we found PKM2 could interact with p27, and decreased level of p27 in PKM2 specific siRNA-transfected astrocyte cells. Thus, PKM2 also might negasitive regulate p27 following SCI and the repair process, and further influence the cyclin E/CDK2 in cell cycle progression.

Cell cycle activation and related proteins expression exert the important function on astrocytes proliferation after CNS injury.¹ The G1/S is a crucial cell cycle checkpoint, which is controlled by 2 cell cycle kinases, CDK4/6-cyclin D1 and CDK2-cyclin E, and the transcription complex that includes Rb and E2F.⁴⁴ Rb phosphorylation by CDK4/6 and CDK2 dissociates the E2 promoter-binding protein dimerization partners (E2F) from the pRb/E2F complex, and dissociated E2F induces transcription of cyclin E1, amplifying the G1- to S-phase switch, and being required for DNA replication.⁴⁵ Consistent with our studies, we show that the expression of CDK4/cyclin D1, CDK2/cyclin E and phosphorylation of Rb were increased parralel with PKM2, indicating cell cycle activation. We also show that knockdown PKM2 by siRNA can decrease p-β-catenin expression and p27 degeneration as well as PCNA expression, further inhibiting over proliferation of astrocytes. Base on the above data, we speculated that PKM2 promoted astrocyte proliferation via regulating p27 and β-catenin signal pathway after spinal cord injury. Recent evidence revealed PKM2 prevented cell apoptosis via modulating Bim stability in hepatocellular carcinoma.⁴⁶ Here we observed PKM2 also expressed in neurons, although expression changes in neurons were not evident. Whether PKM2 could inhibit neurons apoptosis after spinal cord injury should be further explored.

In summary, we have shown that PKM2 expression were positive associated with proliferation index of PCNA after spinal cord injury in adult rat. Our findings also indicate traumatic SCI induce nuclear translocation of PKM2 to interact with β-catenin and p27, and further regulate cyclin proteins in astrocytes expression, which contribute to the dysregulation of cell cycle and precede the progression of astrocyte cells proliferation. Although the detailed biological functions of PKM2 in glial cells proliferation have not been fully identified, our discoveries firstly uncovered PKM2 as a positive regulator of cyclin proteins in spinal cord injury and provided a new potential therapeutic target for CNS injury and repair.

Disclosure of Potential Conflicts of Interest

No potential conflicts of interest were disclosed.

Acknowledgments

This work was supported by the National Natural Science Foundation of China (No. 81271367).